Imperial expansion, public investment, and the longpath of history: China’s initial political unificationand its aftermathHui Fanga, Gary M. Feinmanb,1, and Linda M. Nicholasb

aSchool of History and Culture, Shandong University, Jinan 250100, China; and bScience & Education, Field Museum of Natural History, Chicago, IL 60605

Edited by Linda R. Manzanilla, Universidad Nacional Autónoma de México, México, D.F., Mexico, and approved December 5, 2014 (received for reviewOctober 4, 2014)

The Neolithic (ca. 8000–1900 B.C.) underpinnings of early Chinesecivilization had diverse geographic and cultural foundations in dis-tinct traditions, ways of life, subsistence regimes, and modes ofleadership. The subsequent Bronze Age (ca. 1900–221 B.C.) wascharacterized by increasing political consolidation, expansion,and heightened interaction, culminating in an era of a smallernumber of warring states. During the third century B.C., the QinDynasty first politically unified this fractious landscape, across anarea that covers much of what is now China, and rapidly instituteda series of infrastructural investments and other unifying mea-sures, many of which were maintained and amplified during thesubsequent Han Dynasty. Here, we examine this historical se-quence at both the national and macroscale and more deeply fora small region on the coast of the Shandong Province, wherewe have conducted several decades of archaeological research. Atboth scales, we examine apparent shifts in the governance of localdiversity and some of the implications both during Qin–Han timesand for the longer durée.

China | empire | archaeological settlement patterns | political unification

Two millennia ago, two empires, situated at the opposite endsof Eurasia, dominated their respective regions (1). To the

west, Rome was in the process of expanding its conquests acrossEurope, whereas to the east, the Han, having assumed powerfollowing the short-lived Qin Dynasty, ruled over most of what istoday China. Although at their peaks these polities were roughlycomparable in spatial extent, the political legacies that followedtheir declines were markedly different (2). Whereas the boundsof the Roman Empire were never historically reconstituted,Chinese regions, after the initial Qin unification (221 B.C.),were reintegrated perpetually into one political unit, despiteintermittent periods of disunity (3). From the founding of theempire to its early 20th-century dissolution, the area from theMongolian steppe to the South China Sea was ruled by a singleauthority for roughly half this period (1, 2, 4).Why humans cooperate in large, often relatively durable social

groupings is a key question for contemporary research (5, 6).Thus the repeated historical renegotiation of China’s continent-scale political consolidations remains a scholarly focus aftermore than a generation of attention (3, 4, 7, 8). One perspective(3, 4) places great emphasis on biogeography and defensiveconcerns, specifically the persistent, perceived military threat ofmobile peoples along China’s northern frontier. Althoughnorthern invasions did occur periodically during China’s history(3, 8) and military threat does provide strong incentives forcollective action (3, 9, 10), many questions, however, remainunanswered. Why, in China, was the military challenge met bysuccessful, albeit not always long-lived, political recon-stitutions (threats from mobile peoples had no such collabo-rative effect in Europe) (1)? Why did areas distant and lessthreatened by the northern frontier reintegrate, and why didthe repeated reestablishment of empire reunify a landscapeclose to the limits achieved originally by the Qin? Theseunanswered questions belie a problematic but oft-held premise:

that widespread cooperation was somehow easy to attain inChina, perhaps because the populace shared a long-standingcultural tradition or ethnic identity (11).In contrast, we argue that Chinese collective identities (12),

which became social building blocks of large-scale political in-tegration, were constructed, as evidenced elsewhere (13, 14), andcertainly were not primeval. We recount processes and events,both before and after the initial Qin unification, that underpinnedthe making of more overarching Chinese collective identities,subsuming strong elements of local customary diversity (15). Toassess this sequence, we examine multiple analytical scales. First,we synthesize documentary histories (which generally reflect theperspectives of governing and elite principals) and archaeologicaloverviews to recapitulate the transitional sequence that ran fromdiverse networks of agricultural communities across China duringthe Neolithic (ca. 8000–1900 B.C.) to the Qin unification. Then,to assess the ways in which political edicts and governance practicesplayed out for lives on the ground, we review the findings fromone region of China (the coast of Shandong Province), even-tually conquered and integrated by the Qin–Han, where we areconducting a systemic regional archaeological settlement pat-tern survey (16–18). Through this multiscalar investigation, weevaluate both shifting strategies of governance and the waysthat they were negotiated and received in one local context.

Paths from Village Life to Political UnificationOver the last three decades, the advent of new archaeologicaland archaeobotanical findings has led to a dramatically newunderstanding of the transitions from mobile gatherer–huntersto sedentary life and farming in China (19, 20). Prior models thatenvisioned the radiation of agricultural villages from the Central

Significance

During the Bronze Age, the diverse cultural traditions of theearlier Chinese Neolithic were interconnected into more ex-pansive political and economic networks that culminated inthe establishment of the region’s first empires (Qin and Han).Drawing on documents and archaeology, we outline the pro-cesses associated with this political consolidation and steps thatwere taken during imperial governance at both the nationaland a specific local scale (coastal Shandong Province), where wehave conducted two decades of archaeological settlement pat-tern research. Through the juxtaposition of macro- and micro-scale analysis, we document human impacts of infrastructuralinvestments, interactive technologies, and ideological tenetsthat were implemented during the Qin–Han dynasties, whichhelped establish the rough spatial configuration of what hassince been China.

Author contributions: H.F., G.M.F., and L.M.N. designed research, performed research,analyzed data, and wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.1To whom correspondence should be addressed. Email: gfeinman@fieldmuseum.org.

9224–9229 | PNAS | July 28, 2015 | vol. 112 | no. 30 www.pnas.org/cgi/doi/10.1073/pnas.1419157112

Plain out to the rest of the continent have been revised to reflecta deeper time and more spatially mosaic Neolithic era that sawsignificantly different subsistence preferences, sequences ofchange, community patterns, modes of leadership, and evensymbolic representations across the diverse landscapes of China(21–23). The extent of variation extended well beyond the vari-ability in subsistence regimes that favored rice in the south andvarieties of millet in the north. By 5000 B.C., across China,there was marked diversity in the ways that power was funded,materialized, and communicated between emergent leaders andassociated subalterns with seemingly more collectively orientedgovernance in some regions, such as the Central Plain, and moreruler-centric political formations in others, such as the Yangzidelta (21, 24, 25).By the advent of the second millennium B.C., what is today

China was divided into many small polities that, although inter-acting in down-the-line socioeconomic networks, had distinctiveways of manifesting power and communicating their identitiesthrough the use of diverse ceramic and other artifactual traditions(26). Subsequent textual accounts place the roots of early Chinesecivilization in the Central Plain, principally with the Shang (Fig. 1).Shang rule was aristocratic, based on the veneration of clanancestors, access to metal resources, the production and dis-tribution of bronze vessels, writing, and the materialization ofpowerful symbols. During the second millennium, precociousdevelopments in urbanization, high-intensity metal production,and early writing did occur on the Central Plain (26, 27). Thefirst experiments in metal work for China, however, occurredoutside this area (26, 28), and massive population centers alsoarose elsewhere associated with highly different indigenous cul-tural traditions (26).The initially perceived conception of ancient Shang military

expansionism has narrowed from earlier interpretations, as detailedarchaeological investigations have been undertaken in regions ad-jacent to the Shang heartland (12, 27). Although involved in theacquisition of metals, and so extended networks of trade, Shangconquest and political expansion was limited in geographical extent.Texts do confirm that the Shang viewed their enemies, especiallythose who were able to successfully withstand their military foraysor lay beyond them, as “barbarian” peoples, with distinctive col-lective identities (29). For example, according to Shang oraclebones, the barbarian peoples to the east (in Shandong Province)had their own distinguishing rituals, supernatural spirits, religiousobjects, and ceremonies and sacrifices (29).The defeat of the Shang by the Zhou, whose homeland was in

northwest China (ca. 1046 B.C.), was the first episode of severalduring Chinese history where outsiders invaded and conquered

the Central Plain, but then adopted many of the governing andcultural practices of that region (12, 30). The epoch of Zhou rulewas the longest enduring Chinese dynasty, albeit also a time ofcataclysmic change. Many of the institutions and practices thatunderpinned later notions of Chinese identity and governancefirst were implemented during that era (30, 31), but the mostsignificant of these shifts occurred after the Western Zhou pe-riod (1046–771 B.C.). Western Zhou rulers made near-continuousefforts to vanquish the people of Shandong to the east, and al-though they conquered some areas and disrupted others,people still referenced in texts as “eastern barbarians” wereable to rebuff their military efforts, especially in coastalShandong (29, 31).The subsequent centuries of the Zhou period (Eastern Zhou,

771–221 B.C.) were marked by both political consolidation andsignificant shifts in the principals and practices of governance(12, 30). As typified by the writings of Confucius, there wasa shift in leadership and governance from more aristocraticforms that were legitimized through linear clan-based ties tomore explicit moral codes and defined social expectations forrulers, the ruled, and those who administered the functions ofgovernment (9, 30, 32). An emergent nonhereditary bureaucracyimplemented systems of taxation and legal codes (30, 32). TheEastern Zhou Dynasty endeavored to expand its political hege-mony in all directions through conquest and political alliance,but met perpetual resistance, and in the latter centuries of thisera, centralized authority largely broke down, leaving 5–10 for-merly vassal states across the Chinese landscape to vie for powerand territorial control during the Warring States era (453–221B.C.) (Fig. 2) (15, 30, 33). Growth in the size of polities alongwith the challenges of ruling over increasingly diverse pop-ulations in the face of persistent political and military com-petition likely prompted fiscal reforms and the adoption of thesomewhat more collectively oriented policies and practices,such as investments in public goods (9; ref. 34, pp. 338 and339). As diverse populations were integrated into ever-largerstates during this era, the contrasts expressed in documents ofthe time between Zhou peoples and the barbarians outside thebounds of the empire were sharpened (32). The long Zhou erawas the time when “the Chinese defined for themselves a cul-ture as well as a world” (ref. 35, p. 550).

Qin Reforms, Han Programs, and Shifting Notions of IdentityIn China, there was no collective Chinese identity at the outset ofthe Warring States period; rather, traditional affiliations, in-cluding language, funerary customs, and notions of time (alma-nacs), often were associated with local states, albeit polities now

Fig. 1. China, showing places mentioned in text.

Fig. 2. Polities of the Warring States period and location of the Qi wall.

Fang et al. PNAS | July 28, 2015 | vol. 112 | no. 30 | 9225

ANTH

ROPO

LOGY

SPEC

IALFEATU

RE

growing in size (12). The Qin state, one of these local polities,was vassal to the Zhou and situated to the northwest of thelatter’s core (12, 33). During the fourth century B.C., the Qinbegan an episode of conquest that culminated in China’s politicalunification. Although subsequent Qin imperial rule was exceed-ingly short (221–207 B.C.), the changes set in motion duringthese times contingently underpinned a national identity andthe course of subsequent Chinese history (12, 15, 33, 36).With military conquests, the Qin state instituted new govern-

mental policies. An array of initiatives was designed to reorderlongstanding social relations, thereby severing the ties betweensubalterns and aristocratic local lords. Other reforms were meant tostrengthen the Qin state by increasing its tax base, expanding itsarmies, and redirecting the loyalties and affiliations of all its con-stituents to a national identity, the Qin state (12, 36). The reformswere fundamental and multifaceted, including diminishment ofrituals associated with ancestor veneration, binding households intogroups of five that were intended to share responsibility for eachother’s behavior, naming commoners, and placing greater expec-tations on farmers to join the military (33). Military prowess andgovernmental service became means for subaltern social mobility(33, 36).Later Qin changes fundamentally altered the rhythms of life

and networks of communication. A uniform system of coinagewas introduced. A census was implemented. The written scriptwas unified to enhance communication across language/dialectcommunities. Registries and tolls were imposed to limit andtrack personal movement. Daily and yearly almanacs were syn-chronized, with the impact of harmonizing rituals and com-memorations over broader spatial realms (12, 36).Some Qin directives, both during the century before and im-

mediately after unification, were intended to break down andreorient existing socialized landscapes. Numerous roads werebuilt, and river transport was improved. Efforts were made toconnect the walls that had been built at the northern limits ofthree of the warring states, which became the genesis of China’sGreat Wall (12, 36). After the defeat of the last competing state,the Qi, in eastern Shandong, 30,000 families were moved to the

coast to establish a new regional capital at Langyatai (17; ref. 37,p. 47). Such mass relocations were ordered elsewhere as well (38).Overall, the reforms instituted by the Qin were geared for

persistent war, conquest, and the bureaucratic redefinition of anexpansive domain (39). Qin governance was underpinned by, andin certain respects rejected, prior Zhou statecraft, includingcertain fundamental social contracts and moral codes (33). Al-though early texts state that the Qin rulers recognized that theworld had changed with unification, their policies and practiceswere legitimized and guided by claims that the emperor was uniquewith godlike powers, destined for eternal life (36). Faced with manypotential royal rivals and a massive empire, the Qin Dynasty en-dured only a few years following the death of the first emperor,Qinshihuang (also known as Shihuangdi). Infrastructural invest-ments could not keep up with the rapid tempo of expansion (38).Dynastic power shifted to the Han (206 B.C.–A.D. 220), who

maintained many of the unifying initiatives of their immediatepredecessors but readopted key elements of Zhou statecraft andmorality, including the “Mandate of Heaven,” which proclaimedthe emperor’s place as a product of supernatural authority ratherthan cosmic process (36). During the first centuries of the HanDynasty, although greater degrees of power were shared with localdespots, the imperial authority of the centralized empire was re-affirmed (38). Unity and unification were preferred and associatedwith leadership. The overarching imperial order gave primacy touniversality but did not mandate uniformity, thus integrating ele-ments of diversity (40–42).At the same time, investment continued in many public goods

projects begun by the Qin. These initiatives and others pushedand pulled peasants to intensify farming, thereby reaching pre-viously unprecedented levels of agrarian production (43, 44). Inthe A.D. 2 census, the first preserved, the population of Chinaexceeded 57 million people, which is estimated to be a significantincrease from even a few centuries before (38, 45). The HanDynasty bequeathed political, social, and ideological foundationsfor empire that endured largely intact for two millennia (38).

Fig. 3. Early Longshan settlement in southeastern coastal Shandong.

Fig. 4. Middle Longshan settlement in southeastern coastal Shandong.

9226 | www.pnas.org/cgi/doi/10.1073/pnas.1419157112 Fang et al.

Integrating Diversity: A Microscale Focus from CoastalShandongWe have outlined the sequence of changes at the national scalethat led to the 221 B.C. unification of China and the constructionof belief systems, ideological constructs, bureaucratic blueprints,and socioeconomic connections that served as the basis for laterepisodes of political consolidation. The empirical basis for thisaccount, however, is almost entirely top–down, derived fromtexts, leaving unaccounted for the perspective and presence ofmore than 90% of the population. In this section, we cannot givevoice to that majority, but we can gain insights into how thesemassive big-scale transitions interfaced with their lives on theground for at least one local region, coastal Shandong, that hasbeen systematically investigated through regional archaeologicalsettlement pattern surveys (16–18).Over the past two decades, our collaborative team has im-

plemented a broad-scale walkover of more than 2,288 sq km,covering a large coastal basin immediately south of theShandong Peninsula in eastern China (Fig. 1) (16–18). The aimof the archaeological survey is to record broad-brush spatialand temporal perspectives on changing settlement patterns anddemography for the focal region by the systematic recording ofsurface archaeological remnants of ancient settlement that arerecovered on the ground and diligently entered on maps (ref. 27,pp. 221–222; 46). Pottery, the most abundant artifact recovered,provides a basic means for dating the surface remnants of pasthuman landscape use and therefore ancient settlement pat-terns across time. Other visible architectural and archaeologi-cal features, such as tomb monuments, platforms, ancient walls,and exposed pits, also are recorded systematically (16, 18).Based on the settlement survey, coastal Shandong was not

inhabited by a wide network of farming communities until thelatter half (ca. 3000 B.C.) of the Neolithic period (16, 18). Oncethe area was settled, however, the population grew rapidly, andby the Early Longshan period (ca. 2600 B.C.) the regionalpopulation aggregated around three large settlements—Yaowangcheng in the south, Liangchengzhen in the center, andHetou in the north—with population rather evenly distributedacross the area (Fig. 3). Although the material culture inventory

between these three sectors appears homogeneous, these areaseach have somewhat different topographic/environmental set-tings. The southernmost area is very flat, with wide rivers thatempty into the coast, and is used to farm mostly rice today. Themiddle sector includes somewhat more rolling topography, withnarrower watercourses, and the farming economy is mixed, in-cluding wheat, with only patches of rice. The northernmost area ismost dissected and hilly, although the suite of crops today is broadlysimilar to that immediately south of it, with greater representationof cold-tolerant plants, such as cabbage.By the subsequent Middle Longshan period (ca. 2400 B.C.), the

population in and around Hetou in the north markedly declined asthe two other centers continued to exceed all others in size (Fig.4). The ceramics and other material culture in the coastal basinwere different from that found on the Shandong Peninsula (47),and so we propose that the northern end of our survey region mayhave become a kind of buffer or shatter zone between peopleaffiliated with these different traditions. Likewise, although thesize of these central communities and the density of overall set-tlement on the coast were not at all indicative of mobile “barbar-ians referred to in documents,” the ceramic and other artifactual/symbolic traditions on the coast were sufficiently distinct fromthose found on the Central Plain at this time that they could havebeen viewed by the people on the opposite side of the Shandongmountains as ethnically “other.” Furthermore, in accord with thetexts, we found no artifactual or other material indication ofa Shang invasion or direct, persistent presence on the Shandongcoast early in the Bronze Age (29).Later in the Bronze Age, incursions and influences from the

West did have significant effects on the Shandong coast, al-though at least during the Western Zhou period, we do not thinkthat most of the coastal population was incorporated into ex-pansive polities with capitals outside the investigated region.Nevertheless, by late in the second millennium B.C., the settle-ment patterns on the Shandong coast shifted markedly, as boththe size of the largest centers and the region’s overall populationdeclined (Fig. 5). During the Western Zhou period, many ofthe larger communities were moved to elevated, somewhat

Fig. 5. Western Zhou settlement in southeastern coastal Shandong.

Fig. 6. Eastern Zhou settlement in southeastern coastal Shandong.

Fang et al. PNAS | July 28, 2015 | vol. 112 | no. 30 | 9227

ANTH

ROPO

LOGY

SPEC

IALFEATU

RE

defensible, locations along the western edge of the study region.Together these communities define an alignment of settlementsthat then extends from the western edge of the basin to the coast(north of where Liangchengzhen is situated). The placements ofthese settlements define or demarcate the southern two-thirds ofthe surveyed region from the north and west. Most of the areanorth of this west–east line of settlements continued to besparsely occupied as it had been since the decline of Hetoufollowing the Early Longshan period.By Eastern Zhou times, there was another significant shift

in the settlement pattern of the coastal basin where we areinvestigating (Fig. 6). Many of the earlier Western Zhoudefensible settlements that seemed to define a coastal do-main were abandoned or diminished in size. Yet overall, thecoastal population increased. At the same time, the headtowns or central places in the region remained relativelysmall in size. In accord with the political consolidationestablished in documents, the coastal basin likely was in-corporated into one or more larger states whose capitals werelocated outside the study region. Although the regionalpopulation expanded during Eastern Zhou, the populationalong the northern edge did not, remaining sparse as it wasfor more than a millennium.The Great Wall of the Qi State (seemingly China’s first great

wall), which was erected early in the Warring States era, was builteast-to-west across much of what is today Shandong Province(Fig. 2). The rammed earth wall that in places was lined orbolstered by stones extends more than 600 km. It defined thesouthern limits of the Qi polity, which was the last of the largewarring polities to be engulfed by Qin armies before unification.During the survey, we were able to follow the easternmost ex-tension of the Qi wall for 50 km as it ran across the northernlimits of our study region (Fig. 7). Therefore, the political borderthat the wall demarcated in Warring States times likely had beena kind of boundary for more than 1,500 years. At least for itseasternmost 50 km, the Qi wall was erected along ridge topsand followed the highest contours possible, descending to lowerelevations only when there was no other option. The Qi wallwas not exceedingly tall. Thus the wall likely was most effectiveas a means to slow down the giant military infantries that arereported in textual accounts of the Warring States period (48).Ultimately, after the wall was breached, and the Qi state

defeated by the Qin, Qinshihuang visited these new conquests.He climbed Langya Mountain, overlooking the coast, and erec-ted a stela to proclaim his rule (17, 49, 50). Later, according totextual accounts, as an expression of his delight with the regionand the sea, he relocated 30,000 families to Langyatai (37; ref.38, p. 47). In line, however, with other programs to constructa more unified imperial domain, it seems likely that Qinshihuangwas intent to break down long-lived cultural and political divi-sions by placing a hub of economic and political activity in

an area that for generations was settled sparsely. These Qinreforms, largely reaffirmed by the Han, appear to have achievedtheir aim. The population grew rapidly in the region during Qin–Han, with the greatest expansion in the northern sector, sur-rounding the emergent local capital of Langyatai (Fig. 8). Hansettlements were established in what was the earlier shatter zone,as the harbor and salt resources around the provincial capitalfostered local economic growth (17). Across the coastal basin,the Han population reached levels not achieved earlier. Anumber of towns grew up near natural routes and passes to thewest. These new second-tier settlements likely served as nodalcommunities that linked the coast to more inland areas at a timeof increasing communication and socioeconomic integration.

Concluding Thoughts and ImplicationsBy examining (at two analytical scales) diachronic change inChina from the Neolithic to the nation’s first episode of unifi-cation, we have documented how a diverse landscape inhabitedby people with a multitude of different traditions and economicpursuits were ultimately unified into one of the largest imperialdomains that the world had seen to that date. There was nothingpreordained or strictly biogeographical when these consolidatingtrends first occurred, nor were such factors sufficient when therulers and people of China reintegrated themselves into a singlepolity in the subsequent two millennia. Likewise, variation indialect, culinary traditions, and other customs were maintained,but these elements of diversity did not crystalize into long-standing episodes of political fragmentation (51). Rather, theglobally unique tendency for China to be politically unified sorepeatedly at a near continental scale (1–4) was in large part theconsequence of the social construction of an amalgam of broadlyheld political ideals, institutional structures and ties, unifiedcommunication technologies and commerce networks (ref. 52,p. 23), and collective traditions and memories that were initiallynegotiated and adopted during the Shang, Zhou, Qin, and Haneras. Although these practices and institutions certainly did notremain entirely stagnant, they did become part of the fabric of

Fig. 7. Survey crew walking on the Qi wall as it winds its way up the hilltopin front of them.

Fig. 8. Qin–Han settlement in southeastern coastal Shandong.

9228 | www.pnas.org/cgi/doi/10.1073/pnas.1419157112 Fang et al.

Chinese collective identities underpinning subsequent efforts forthe reformulation and sustenance of political unity.More specifically, the forging of new moral codes and so-

ciopolitical contracts and institutions during the Bronze Agethat reworked relations between states, elites, and ordinarycitizens was a foundation and catalyst for subsequent and re-peated episodes of political consolidation (34, 39). In Europeduring a comparable period of political expansion, Romangovernance, with mostly smaller scale enemies and a focus onresource extraction, did not institute or negotiate comparableshifts in these fundamental interpersonal relations (ref. 1, pp. 16and 17; 53). During the Roman Empire, patronage and patri-monialism remained key pillars of imperial power (1), and thelegacy after imperial collapse left a much more culturally andpolitically fragmented landscape. As a consequence, subsequent

military threats that required defensive responses served as spursfor more collective action and political response in China,whereas parallel threats still today do not have a comparableeffect at the opposite end of the Eurasian land mass.

ACKNOWLEDGMENTS. We thank all the national, provincial, local, andShandong University officials who made our project possible and assisted usin many ways over the years. Many colleagues and students participated inthe study; although we cannot list them all, their efforts are appreciated.Linda Nicholas and Jill Seagard prepared the illustrations. Linda Nicholastook the photograph. We gratefully acknowledge the financial supportreceived from the National Science Foundation, Henry Luce Foundation,Wenner-Gren Foundation, Field Museum, Tang Foundation, Chinese Pro-gram to Introduce Disciplinary Talents to Universities (111-2-09), QingdaoCity Institute (Shandong, China), and Chinese Project III to Probe the Originof Civilization.

1. Scheidel W (2009) From the “Great Convergence” to the “First Great Divergence.”Rome and China: Comparative Perspectives on Ancient World Empires, ed Scheidel W(Oxford Univ Press, New York), pp 11–23.

2. Ko CY, Koyama M, Sng TH (2014) Unified China and Divided Europe. Available at ssrn.com/abstract=2382346. Accessed September, 20, 2014.

3. Turchin P (2009) A theory for the formation of large empires. Journal of Global His-tory 4(2):191–217.

4. Ko CY, Sng TH (2013) Regional dependence and political centralization in imperialChina. Eurasian Geography and Economics 54(5-6):470–483.

5. Kennedy D, Norman C (2005) What don’t we know? Science 309(5731):75.6. Carballo DM, Roscoe P, Feinman GM (2014) Cooperation and collective action in the

cultural evolution of complex societies. J Archaeol Method Theory 21(1):98–133.7. Lattimore O (1967) Studies in Frontier History: Collected Papers 1928-1958 (Mouton, Paris).8. Barfield TJ (1989) The Perilous Frontier: Nomadic Empires and China (Blackwell, Oxford).9. Blanton RE, Fargher L (2008) Collective Action in the Formation of Pre-Modern States

(Springer, New York).10. Turchin P (2006)War and Peace andWar: Life Cycles of Imperial Nations (Pi Press, New York).11. Eisenstadt SN, Giesen B (1995) The construction of collective identity. Europ J of Sociol

36(1):72–102.12. Yates RDS (2001) Cosmos, central authority, and communities in the early Chinese em-

pire. Empires: Perspectives from Archaeology and History, eds Alcock SE, D’Altroy TN,Morrison KD, Sinopoli CM (Cambridge Univ Press, Cambridge, UK), pp 351–368.

13. Nagel J (1994) Constructing ethnicity: Creating and recreating ethnic identity andculture. Soc Probl 41(1):152–176.

14. Emberling G (1997) Ethnicity in complex societies: Archaeological perspectives. J ArchaeolRes 5(4):295–344.

15. Pines Y, Falkenhausen LV, Shelach G, Yates RDS (2014) General introduction: Qin historyrevisited. Birth of an Empire: The State of Qin Revisited, eds Pines Y, Falkenhausen LV,Shelach G, Yates RDS (University of California Press, Berkeley, CA), pp 1–34.

16. Underhill AP, et al. (2008) Changes in regional settlement patterns and the de-velopment of complex societies in southeastern Shandong, China. J Anthropol Ar-chaeol 27(1):1–29.

17. Feinman GM, Nicholas LM, Hui F (2010) The imprint of China’s first emperor on thedistant realm of eastern Shandong. Proc Natl Acad Sci USA 107(11):4851–4856.

18. Fang H, et al. (2012) Archaeological Report of Regional Systematic Survey in theSoutheast Shandong (Science Press, Beijing). Chinese.

19. Cohen DJ (2011) The beginnings of agriculture in China: A multiregional view. CurrAnthropol 52:S273–S293.

20. Zhao Z (2011) New archaeobotanic data for the study of the origins of agriculture inChina. Curr Anthropol 52:S295–S306.

21. Liu L (2004) The Chinese Neolithic: Trajectories to Early States (Cambridge Univ Press,Cambridge, UK).

22. Falkenhausen LV (2007) Review of The Chinese Neolithic: Trajectories to Early States.Harv J Asiat Stud 67(1):178–193.

23. Bettinger RL, Barton L, Morgan C (2010) The origins of food production in NorthChina: A different kind of agricultural revolution. Evol Anthropol 19(1):9–21.

24. Li X (2013) The later Neolithic period in the central Yellow River valley area. A Com-panion in Chinese Archaeology, ed Underhill AP (Blackwell, Chichester, UK), pp 213–235.

25. Qin L (2013) The Liangzhu culture. A Companion in Chinese Archaeology, edUnderhill AP (Blackwell, Chichester, UK), pp 574–596.

26. Bagley R (1999) Shang archaeology. The Cambridge History of Ancient China: Fromthe Origins of Civilization to 221 BC, eds Loewe M, Shaughnessy EL (Cambridge UnivPress, Cambridge, UK), pp 124–231.

27. Liu L (2009) State emergence in early China. Annu Rev Anthropol 38:217–232.28. Yao A (2010) Recent developments in the archaeology of southwestern China.

J Archaeol Res 18(3):203–239.29. Fang H (2013) The eastern territories of the Shang and Western Zhou: Military ex-

pansion and cultural assimilation. A Companion in Chinese Archaeology, edUnderhill AP (Blackwell, Chichester, UK), pp 473–492.

30. Falkenhausen LV (2006) Chinese Society in the Age of Confucius (1000-250 BC): The

Archaeological Evidence (Cotsen Institute, Los Angeles).31. Feng L (2006) Landscape and Power in Early China: The Crisis and Fall of the Western

Zhou 1045-771 BC (Cambridge Univ Press, Cambridge, UK).32. Falkenhausen LV (1999) The waning of the Bronze Age: Material culture and social

developments. The Cambridge History of Ancient China: From the Origins of Civili-

zation to 221 BC, eds Loewe M, Shaughnessy EL (Cambridge Univ Press, Cambridge,

UK), pp 450–544.33. Shelach G, Pines Y (2006) Secondary state formation and the development of local

identity: Change and continuity in the State of Qi (770-221 BC). Archaeology of Asia,

ed Stark MT (Blackwell, Malden, MA), pp 202–230.34. Deng KG (2012) The continuation and efficiency of the Chinese fiscal state, 700 BC-AD

1911. The Rise of Fiscal States: A Global History, 1500-1914, eds Yun-Casalilla B,

O’Brien PK (Cambridge Univ Press, Cambridge, UK), pp 335–352.35. Hsu CY (1999) The Spring and Autumn period. The Cambridge History of Ancient

China: From the Origins of Civilization to 221 BC, eds Loewe M, Shaughnessy EL

(Cambridge Univ Press, Cambridge, UK), pp 545–586.36. Lewis ME (2007) The Early Chinese Empires: Qin and Han (Harvard Univ Press,

Cambridge, MA).37. Sima Q (1993) Qin Dynasty, Records of the Grand Historian, trans Watson B (Columbia

Univ Press, New York), Rev Ed, Vol 1.38. Bodde D (1986) The state and empire of Ch’in. The Cambridge History of China,

Volume 1: The Ch’in and Han Empires, 221 BC-AD 220, eds Twitchett D, Loewe M

(Cambridge Univ Press, Cambridge, UK), pp 20–102.39. Kiser E, Cai Y (2003) War and bureaucratization in Qin China: Exploring an anomalous

case. Am Sociol Rev 68(4):511–539.40. Loewe M (1994) China’s sense of unity as seen in the early empires. Toung Pao 80(1/3):

6–26.41. Lewis ME (2006) The Construction of Space in Early China (State Univ of New York

Press, Albany, NY).42. Pines Y (2005-2006) Bodies, lineages, citizens, and regions: A review of Mark Edward

Lewis’ The Construction of Space in Early China. Early China 30:155–188.43. Bray F (1979-1980) Agricultural technology and agrarian change in Han China. Early

China 5:3–13.44. Anderson EN (1989) The first green revolution: Chinese agriculture in the Han Dy-

nasty. Food and Farm: Current Debates and Policies, eds Gladwin C, Truman K (Univ

Press of America, Lanham, MD), pp 135–151.45. Duan CQ, Gan XC, Wang J, Chien PK (1998) Relocation of civilization centers in an-

cient China: Environmental factors. Ambio 27(7):572–575.46. Kowalewski SA (2008) Regional settlement pattern studies. J Archaeol Res 16(3):

225–285.47. Luan F (1997) On phases and regional types of the Haidai Longshan culture. Collected

Works on the Archaeology of the Haidai Area, ed Luan F (Shandong Univ Press, Jinan,

China), pp 229–282. Chinese.48. Lewis ME (1999) Warring states: Political history. The Cambridge History of Ancient

China: From the Origins of Civilization to 221 BC, eds Loewe M, Shaughnessy EL

(Cambridge Univ Press, Cambridge, UK), pp 587–650.49. Kern M (2007) Imperial tours and mountain inscriptions. The First Emperor, ed Portal J

(British Museum Press, London), pp 105–113.50. Sima Q (1993) Han Dynasty, Records of the Grand Historian, trans Watson B

(Columbia Univ Press, New York), Rev Ed, Vol 2.51. Skinner GW (1985) Presidential address: The structure of Chinese history. J Asian Stud

44(2):271–292.52. Skinner GW (1977) Introduction: Urban development in imperial China. The City in

Late Imperial China, ed Skinner GW (Stanford Univ Press, Stanford, CA).53. Ertman T (1997) Birth of the Leviathan (Cambridge Univ Press, Cambridge, UK).

Fang et al. PNAS | July 28, 2015 | vol. 112 | no. 30 | 9229

ANTH

ROPO

LOGY

SPEC

IALFEATU

RE